Artefact MCP Server is an open-source Model Context Protocol (MCP) server purpose-built for B2B go-to-market revenue intelligence. It enables AI assistants — including Claude Desktop, Claude Code, and Cursor — to read, analyze, and act on GTM strategy data through natural language, without requiring manual data translation or custom dashboard builds. The product implements Anthropic's Model Context Protocol, a standard that allows AI agents to call structured tools via conversational language. Artefact MCP exposes seven tools that cover the core functions of a revenue intelligence stack: pipeline signal detection, constraint analysis, value engine analysis, GTM commit drafting, ICP triangulation, RFM analysis, and pipeline health scoring. Pipeline Signal Detection classifies pipeline activity into six signal types — Momentum Shift, Stall Pattern, Conversion Anomaly, Engagement Spike, Risk Indicator, and Expansion Signal — turning raw CRM data into actionable intelligence. The Constraint Analysis tool applies the Revenue Formula to identify the dominant scaling bottleneck across traffic, conversion, deal size, or velocity. The ICP Triangulation Framework goes beyond standard firmographics by scoring prospects across three dimensions: company attributes, behavioral signals, and growth indicators. The RFM Analysis Engine segments customers across 11 categories from Champions to Lost, with retention strategies built in for each segment. Pipeline Health Scoring generates a 0–100 score by analyzing velocity, stage distribution, conversion rates, and deal aging. The product is installed via a single command — pip install artefact-mcp — and requires no API key to run on built-in sample data. Live HubSpot CRM data is available on paid plans. The Free tier includes all seven tools. Pro ($149/month) adds live HubSpot integration, custom RFM thresholds, and custom ICP property mapping. Enterprise ($499/month) adds multi-CRM support, dedicated onboarding, custom signal configurations, and an SLA guarantee. Artefact MCP is published on the MCP Registry, available on PyPI, and licensed under BSL 1.1, converting to MIT in 2030. The source code is maintained on GitHub.